CN111513726B - System for evaluating AMS risk based on IHT dynamic performance - Google Patents
System for evaluating AMS risk based on IHT dynamic performance Download PDFInfo
- Publication number
- CN111513726B CN111513726B CN202010264555.2A CN202010264555A CN111513726B CN 111513726 B CN111513726 B CN 111513726B CN 202010264555 A CN202010264555 A CN 202010264555A CN 111513726 B CN111513726 B CN 111513726B
- Authority
- CN
- China
- Prior art keywords
- ams
- iht
- data
- dsi
- spo
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/14542—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue for measuring blood gases
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
- A61B5/4809—Sleep detection, i.e. determining whether a subject is asleep or not
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
- A61B5/4812—Detecting sleep stages or cycles
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
- A61B5/4815—Sleep quality
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Public Health (AREA)
- Surgery (AREA)
- Veterinary Medicine (AREA)
- General Health & Medical Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Signal Processing (AREA)
- Psychiatry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Anesthesiology (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Optics & Photonics (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention provides a system for evaluating AMS risk based on IHT dynamic performance, which comprises a data acquisition module, a data processing module, a robust enhancement module and an AMS severity monitor, wherein the data acquisition module is used for acquiring data; the data acquisition module acquires SpO from a plurality of continuous IHT periods2Information; data processing module pair obtained SpO2Processing abnormal values, filtering and dividing data segments of the information, and selecting low-oxygen data segments; the index acquisition module acquires a dynamic blood oxygen saturation index DSI of the hypoxia data segment, wherein the DSI is SpO2Duration of holding above a set ratio, denoted as τi(ii) a Robust enhancement module for the dynamic blood oxygen saturation index tauiCarrying out robustness enhancement to obtain an index taur(ii) a AMS severity monitor based on a predetermined reference valueAnd said τrAnd comparing to realize evaluation on the AMS. The invention provides a novel AMS (automatic reference system) prediction index DSI (dynamic stress indicator) system by utilizing a dynamic process of human hypoxia stress reaction reflected by continuous physiological information in IHT (IHT).
Description
Technical Field
The invention relates to a system for evaluating AMS risk based on IHT dynamic performance, and belongs to the technical field of prediction and evaluation of acute altitude diseases.
Background
Acute altitude sickness (AMS) is a particular manifestation of inadequate hypoxia tolerance. This common high altitude reaction condition is caused by the rarefied oxygen and low pressure. Sudden entry into the plateau by a person who is not adapted to the experience of a high-altitude environment can produce related symptoms including headache, insomnia, anorexia, and the like. Although acute altitude disease as a hypoxic stress response helps the body to adapt to changing environments and establish new homeostasis, the sustained response severely affects people's lives in high altitude areas.
In the field of sports, Intermittent Hypoxic Training (IHT) is a very popular training method, which can improve athletic performance of athletes and also alleviate symptoms of acute altitude diseases. IHT adopts a method of intermittently taking low-concentration oxygen to simulate the environment of plateau areas. The method intermittently provides sufficient normal oxygen for body without injury to body. This approach facilitates the establishment of a hypoxia adaptation mechanism, accelerates the adaptation process and reduces the risk of AMS. However, the prediction and evaluation of acute altitude diseases by IHT still face many challenges:
1) the training schemes are various and the results are different. Currently, many studies discuss the effectiveness of IHT and devise training schemes and systems. For example, the intermittent hyperhypoxic training system proposed in document [1] (wanbosu, olygger thanks glazakhstaff, chenkuai, guo, huxiazhou. intermittent hyperhypoxic training system [ P ]. CN110575596A, 2019-12-17.). However, these training protocols differ greatly in hypoxia duration and oxygen concentration, and the experimental conclusions drawn from different IHT studies also differ. Document [2] (T.O' Connor, G.Dubowitz, P.E.Bickler.pulse oxidation in the diagnosis of acid methyl acetic acid assay [ J ]. High Altitude Medicine & Biology,2004,5(3),341 & 348) and document [3] (S.Grant S, N.MacLeod, J.W.Kay, M.Watt, S.Patel, A.Paterson, A.Peacock.Sea. level and acid responses to a. do the y compression physiological responses and acid methyl acetic acid assay kit [ J ]. weigh Journal of Spur, 2002,36 (2)), the results of the same type of assay were analyzed using the same criteria, but opposite criteria. Generally speaking, high-intensity hypoxic training results in poor body adaptation, increasing the risk of acute altitude disease, while low-intensity training may result in less than satisfactory training.
2) The subjectivity of the existing diagnosis method of the acute altitude disease. The Louis Lake Score (Lake Louis Score, LLS) is considered an effective method for diagnosing AMS. LLS investigates the severity of five symptoms (including headache, gastrointestinal symptoms, fatigue or weakness, dizziness or vertigo, and difficulty in falling asleep) in a subject by evaluating questionnaires with symptom scoring scales. According to the internationally widely used criteria, subjects with an LLS score greater than 3 were diagnosed with acute altitude disease. However, this diagnostic method is subjective and will be misdiagnosed if the tester denies the associated symptoms.
3) Static physiological indices are difficult to predict accurately. In addition to the most common LLS, some static physiological indicators can also evaluate AMS. In document [4 ]](M.Burtscher,M.Philadelphy,H.Gatterer,J.Burtscher,M.Faulhaber,W.Nachbauer,and R.Likar,Physiological Responses in Humans Acutely Exposed to High Altitude(3480m):Minute Ventilation and Oxygenation Are Predictive for the Development of Acute Mountain Sickness[J].High Altitude Medicine&Biology,2019,20(2),192-2) The physiological indicators considered to be the most promising candidates for AMS prediction. The Heart Rate (HR) and its variation (HVR) were also used to assess AMS. However, the static index information is highly refined and difficult to reflect dynamic processes.
4) The conventional prediction method is difficult to accurately predict AMS in real time. Document [5] (zhanshui, wikunlun, liuchunxiang, muxin, Yanshi, Zhao Xiaojing, Shijinlong, Jiazhilong, Jiaqian. a plateau adaptability evaluation method and system [ P ]. CN110335678A,2019-10-15.) utilize sample data and doctor experience to establish a plateau adaptability deep neural network evaluation model. The method is influenced by factors such as training samples and the like, and the accuracy is difficult to guarantee. Document [6] (yellow river, Gaoyuan, Liubao, Xugang, Sun Binda.) A kit for predicting the risk of acute mountain sickness by combining four microRNA biomarkers [ P ]. CN107058471A,2017-08-18 ]) utilizes the biomarkers to predict the risk of AMS, and the risk of AMS is difficult to monitor in real time.
The existing indexes for evaluating the severity of the acute altitude disease mainly comprise a subjective index LLS and a static physiological index SpO2And the like. LLS is currently the urgency of international recognitionThe diagnosis standard of the sexual altitude disease is widely applied, but is limited by long evaluation period and subjectivity, and is difficult to diagnose and evaluate accurately in time. Static physiological indicators, but containing highly refined hypoxic stress response information, are susceptible to disturbances that are difficult to avoid in experiments. Subject to differences in experimental design, subject selection criteria, poster height, etc., it is also possible to reach opposite diagnostic conclusions. Therefore, the dynamic process of human hypoxia stress response reflected by the continuous physiological information of IHT is utilized, the index which effectively reflects the severity of acute altitude diseases is designed, and the method for constructing real-time monitoring and performance evaluation has important significance for AMS prediction.
Disclosure of Invention
The invention aims to overcome the defects of the existing indexes and provide a system for evaluating the risk of AMS (advanced coronary artery disease) based on IHT (acute respiratory syndrome) dynamic performance.
The invention is realized by the following technical scheme:
a system for evaluating AMS risk based on IHT dynamic performance comprises a data acquisition module, a data processing module, a robust enhancement module and an AMS severity monitor; wherein the content of the first and second substances,
a data acquisition module for acquiring SpO from multiple consecutive IHT periods2Information;
a data processing module for processing the acquired SpO2Processing abnormal values, filtering and dividing data segments of the information, and selecting low-oxygen data segments;
an index obtaining module, configured to obtain a dynamic blood oxygen saturation index DSI of the hypoxia data segment, where the DSI is SpO2Duration of holding above a set ratio, denoted as τi;
A robust enhancement module for the dynamic blood oxygen saturation indicator tauiCarrying out robustness enhancement to obtain an index taur;
AMS severity monitor based on a predetermined reference valueAnd said τrMake a comparison to satisfyThe people who train are qualified and suitable for entering the plateau area to meetTo be further trained and observed, to thereby enable assessment of AMS risk.
Further, the abnormal value of the present invention is processed as follows: and processing by adopting a method of replacing the abnormal value with a non-zero data average value before and after the abnormal value.
Further, the criteria for selecting hypoxic data segments according to the invention are: first, the length of the hypoxic data segment is greater than a preset valueminSecond, remove the average SpO of the last n samples2Hypoxic segments at levels greater than 90%.
Further, the preset l of the present inventionminIs 250 seconds, and the value of n is 200 seconds.
Further, the dynamic blood oxygen saturation index tau of the hypoxia data segment of the inventioniIs SpO2For a duration of more than 85%.
Further, the robustness of the invention is enhanced as follows:
ordering DSI sequence collected in one IHT period according to size, namely { tauiI 1.. lambda., lambda } where lambda represents the number of suitable hypoxic data segments when i is equal to 1<j is, τi≤τj(ii) a DSI acquired in one IHT period is defined as a sequence { tauiThe twenty-th percentile of i 1, λ, is denoted as τr。
Further, the present invention should be taken as { τ [ ]iWhen the base number of i ═ 1.. lambda. } is smaller than a set value, { τ } is estimated by the following methodiI 1, a second of λA ten percentile; the method specifically comprises the following steps:
let τ beiHas a maximum value ofMinimum value ofWill tauiRange of (1)Divided into sub-intervals of length l of delta segment
Wherein, pij(j ═ 1, 2.. multidot., δ) is τ i1,2, λ falls in the jth subinterval, the twentieth locus falls in the subinterval z*Internal;
s.t.,g(z)≥0
whereinThe area of the first 20% of the total histogram area and the first z-1 segmentThe difference between the two;
let Pi 00, the DSI of a hypoxic training period is defined as τr
τr:=τ+lτ
Further, the present inventionThe risk assessment module assesses: reference valueLLS is used as a classification label and is obtained according to the maximum classification accuracy criterion.
Advantageous effects
The invention provides a novel AMS (automatic reference System) prediction index DSI (dynamic stress indicator) by utilizing a dynamic process of human hypoxia stress reaction reflected by continuous physiological information in IHT (IHT) under the framework of process monitoring and performance evaluation. The present invention also determines criteria for classifying the hypoxic environment-adapted and non-hypoxic environment-adapted training subjects using DSI.
(1) The most outstanding characteristic of the AMS prediction index DSI is to introduce the thought of process monitoring and performance evaluation in the control theory field and analyze SpO in IHT2And continuously monitoring information, and obtaining an AMS prediction index by using the dynamic characteristics of the information. Compared with static indexes, the method contains more abundant information and can realize real-time monitoring and evaluation.
(2) The method provided by the invention utilizes methods such as filtering and the like to enable the provided index to have certain robustness on disturbance such as measurement noise, uncertainty factors in a training process and the like.
(3) According to the invention, the individual does not need to enter a plateau environment, the DSI can also predict the severity of the acute altitude sickness, and the convenience and flexibility of actual monitoring are enhanced.
Drawings
FIG. 1 is a block diagram of a dynamic monitoring and performance evaluation system for acute altitude disease prediction based on intermittent hypoxia training according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a data preprocessing method according to an embodiment of the present invention;
FIG. 3 is a diagram of the low oxygen data segment DSI according to an embodiment of the present invention;
FIG. 4 is a diagram of a DSI robustness enhancement algorithm for an IHT period according to an embodiment of the present invention;
FIG. 5 is a least squares fit plot of DSI metrics of all trainees and physiological metrics measured in the hypoxic chamber in an embodiment of the invention;
FIG. 6 is a DSI comparison before and after intermittent hypoxia training in an embodiment of the present invention;
FIG. 7 is a least squares fit graph of DSI indicators of an intermittent hypoxia training susceptible person and physiological indicators measured in a hypoxic chamber in an embodiment of the invention.
Detailed Description
The invention is described in further detail below with reference to the following detailed description and accompanying drawings: the present example is carried out on the premise of the technical solution of the present invention, but the scope of the present invention is not limited to the following embodiments.
The design idea of the invention is as follows: AMS severity is researched and predicted by utilizing physiological data dynamic characteristics of IHT continuous monitoring process, and SpO obtained from IHT2The information is subjected to data preprocessing (including abnormal value processing, filtering and data segment division) to obtain a Dynamic blood oxygen saturation indicator (Dynamic SpO) of each hypoxia period2Index, DSI) and robustness enhancement, using this Index to predict the risk of the trainee to develop AMS, and finally, designing an AMS severity monitor (classifier) for determining whether the individual is suitable for the plateau environment.
A system for evaluating AMS risk based on IHT dynamic performance comprises a data acquisition module, a data processing module, a robust enhancement module and an AMS severity monitor; as shown in fig. 1, in which,
a data acquisition module for acquiring SpO from multiple consecutive IHT periods2And (4) information.
In this embodiment, for training and pre-adaptation, the data acquisition module acquires SpO from multiple consecutive IHT cycles2Information, { s { S }j(k):k=1,...,ljRepresents the SpO obtained in the j-th IHT period2The original data sequence of (1).
A data processing module for processing the acquired SpO2And carrying out abnormal value processing, filtering and data segment division on the information, and selecting a low-oxygen data segment.
Abnormal values caused by external environments such as instrument and equipment in training are reasonably screened and removed, and high-frequency noise of a sampling signal is filtered. Meanwhile, in the IHT, an oxygen concentration in the form of a periodic square wave is supplied. SpO trainer2The level fluctuates with fluctuations in oxygen concentration. Since the present example focuses on SpO at low oxygen concentrations2The high concentration oxygen section data is removed and only the low concentration oxygen section data is studied. Has a length ofIs defined asA data segment of low oxygen concentration. The last step in the data segment partitioning is to select data segments that meet the following criteria:
the first criterion specifies that the length of the hypoxic data segment is greater than lminShort length data segments are difficult to adequately describe hypoxic stress response. Second criterion removes average SpO of the last n sampling points2Hypoxic segments at levels greater than 90% thus rule out data aberrations caused by deep breathing. To avoid confusion, the number of suitable hypoxic data segments is represented by λ.
An index obtaining module, configured to obtain a dynamic blood oxygen saturation index DSI of the hypoxia data segment, where the DSI is SpO2Duration of holding above a set ratio, denoted as τi。
To assess the capacity of the human hypoxia stress response system, the present embodiment is based on the rise time TrThe concept of (a) proposes an index. The rise time being a deterministic control performance evaluationAnd the index reflects the response speed of the dynamic system following the step input. Consider a dynamic system
Where u is the external input, x is the system internal state, and y is the output. Assuming the system can reach steady state, the rise time TrDefined as the time it takes for the first change in the system response y (t) from 10% to 90% of the steady state value. The shorter the rise time, the faster the system will respond to the step input.
In IHT, the oxygen concentration is used as input u, SpO2The hypoxia stress response system with level as output y is asymptotically stable. In IHT, the oxygen concentration varies with a pre-designed periodic square wave, and once the oxygen concentration decreases, the SpO of the human body2The level is reduced until the SpO in the simulated plateau environment is reached within a certain time2A steady state value. Although there are physiological differences in humans, the mean steady state value fluctuates at a level of 75% to 80%. Is denoted as τiI.e. a reasonable hypoxic data segmentThe DSI of (1). In an IHT hypoxia data segment, tauiIs SpO2For a duration of more than 85%. Tau isiIs defined mathematically as
Sampling period of Ts。
A robust enhancement module for the dynamic blood oxygen saturation indicator tauiCarrying out robustness enhancement to obtain an index taur。
SpO2High sensitivity and low accuracy of (2) are prone to erroneous predictions. Therefore, it is necessary to enhance the robustness of the proposed index. The invention sorts DSI sequences collected in an IHT period according to size, namely { tauiI 1.. lambda.l.. When i is<j is, τi≤τj. DSI of one IHT period is defined as the sequence { tau }iI-the twentieth percentile of 1. If { τiWhen the base of i 1.. lambda.. is smaller than a predetermined value (e.g., smaller than 10), we estimate { τ using the following methodiI-the twentieth percentile of 1.
From the viewpoint of robustness, the estimation method is based on tauiThe area of the frequency distribution histogram of (4) is obtained. Let τ beiHas a maximum value ofMinimum value ofWill tauiRange of (1)Divided into sub-intervals of length l of delta segment
πj(j ═ 1, 2.. multidot., δ) is τiI 1,2, λ falls in the jth subinterval. The twentieth percentile falls within the subinterval z*And (4) the following steps.
s.t.,g(z)≥0
WhereinFor the first 20% of the total histogram area and the first z-1 segmentArea ofThe difference between them. To facilitate the calculation, let π 00. Finally, the DSI of a hypoxic training period is defined as τr
τr:=τ+lτ
If it is notTraining for evaluating an IHT cycle may be less accurate and selected closerτThe value of (A) will also be influenced by SpO2In the vicinity of 85%, the present embodiment calculates DSI of one IHT period by using the above method, in consideration of these two reasons, thereby enhancing the robustness of the index. This process is schematically illustrated in fig. 4.
AMS severity monitor based on a predetermined reference valueAnd said τrMake a comparison to satisfyThe people who train are qualified and suitable for entering the plateau area to meetIs classified as an unadapted individual for further training observation.
Specifically, the method comprises the following steps: according to the reference valueAMS severity monitors (classifiers) were designed. This monitor (classifier) is used to determine whether an individual is in a high altitude environment during IHT. Satisfy the requirement ofIs well-trained and suitable for entering plateau areas, howeverIs classified as an unadapted individual for further training observation. Selection by DSI relationship to LLSLLS is used as a classification label and corresponding classification benchmark is obtained according to the maximum classification accuracy criterionThis criterion was used as a basis to classify people with AMS from those who do not.
The following example demonstrates the correlation of IHT data with AMS indicators, and the IHT data can be processed to detect AMS.
First, 18 subjects were subjected to intermittent hypoxic training. Ten days of continuous intermittent hypoxic training are carried out in a day period (namely, 10 intermittent hypoxic training periods are total). During each training period, the subjects ingested periodically fluctuating concentrations of oxygen, including a low oxygen concentration phase (11% -14% oxygen) lasting 5 minutes and a high oxygen concentration phase (35% -38%) lasting 3 minutes. A high oxygen deficiency device (GO2Altitude hypoxinator) (Biomedtech Australia pty. ltd., Melbourne, Australia) may be employed for intermittent hypoxia training. The hypoxia training device is a mask type device and can provide oxygen with low concentration and high concentration according to the gas separation principle. The device is equipped with a pulse oximeter unit for measuring Heart Rate (HR) and blood oxygen saturation (SpO)2) Time of sampling TsIs 1 second.
Second step, collected SpO2And data preprocessing, including abnormal value processing, filtering and data partitioning.
At SpO2In monitoring, when the sensor does not collect information, "s (k) ═ 0" appears in successive data points "The case (1). This results in the generation of outliers in the original data set s (k), as shown in fig. 2 (a). This embodiment deals with outliers by replacing them with an average of non-zero data before and after the outliers.Defined as the ith exception fragment. When k ∈ { α ∈ [ ]i,βiH, smoothing the data sequence s0(k) Is defined as
For normal data points (s (k) ≠ 0), let s0(k) S (k). The effect of processing the abnormal data is shown in fig. 2 (b).
After processing the abnormal value, the influence of the high-frequency noise is reduced by using a moving average filtering method, and the window width which is the data width participating in the moving average is 50. Length of lfL- ω +1 post-filtering sequence sf(k) Is composed of
The filtering effect is shown in fig. 2 (c).
Has a length ofIs defined asThe data segment division criterion is that the length of the hypoxia data segment is more than 250 seconds and the average SpO is 200s after the hypoxia data segment is removed2Hypoxic segment with a level greater than 90%, namely:
the first criterion specifies that the length of the hypoxic data segment is greater than lminShort length data segments are difficult to adequately describe hypoxic stress response. Second criterion removes average SpO of the last n sampling points2Hypoxic segments at levels greater than 90% thus rule out data aberrations caused by deep breathing. To avoid confusion, the number of suitable hypoxic data segments is represented by λ. The data segment division effect is shown in fig. 2 (d).
And thirdly, obtaining DSI trained every day and enhancing the robustness of the DSI.
First, the DSI, i.e., τ, of each hypoxic period segment is determinediAs shown in fig. 3.
s.t.,ηi(k)≤0.85
The robustness of DSI of one IHT period is evaluated and the DSI sequence { tau } of the period is calculatediThe twenty-th percentile of i 1.., λ } is used as the period DSI. If { τiI 1.. λ } is small, we use the radix in terms of τiIs estimated by the area of the frequency distribution histogram of (1) { tau }iI-1.., λ } in which the histogram subintervals are divided into 15 segments, the calculation of this method is schematically shown in fig. 4.
And fourthly, before and after the hypoxia training, carrying out a hypoxia tolerance test by using the same mask hypoxia training device and obtaining a Hypoxia Tolerance Index (HTI). HTI is SpO of training subjects within 420 seconds2For a duration of more than 85%. For simplicity, with HTI1And HTI2Representing the level of hypoxia tolerance before and after IHT. Defining the difference in the level of hypoxia tolerance before and after training, Δ HTI ═ HTI2-HTI1。
Fifth, after the last IHT cycle, the subjects were sent to a low atmospheric pressure at 9pmThe oxygen chamber (air force medical research institute, beijing, china) was used for overnight sleep experiments. The atmospheric hypoxic chamber simulates the atmospheric environment (oxygen concentration is about 12%) in the area with the altitude of 4300 m. The sleep index includes sleep time, deep sleep time and ratio, night wake time, and physiological index such as blood oxygen saturation (SpO)2) And Heart Rate (HR). The micro-motion sensitive mattress sleep detection system RS-611 (emerging Yang Sheng technology Co., Ltd., Beijing, China) is used for sleep monitoring. The system can make accurate measurements without affecting normal sleep. The next day, 7am, the training subjects were subjected to the Louisis Lake (LLS) test.
Sixthly, by using LLS as a classification label and according to the maximum classification accuracy criterion, the corresponding data can be obtained from the data in the first sub-graph in FIG. 7Is 210 seconds.
This example analyzed the correlation between AMS prediction index derived from IHT data and physiological index measured in hypoxic chamber, the analysis results are shown in Table 1, and Table 1 shows the correlation between DSI index of all trainers in this example and physiological index measured in hypoxic chamber.
Table 1: DSI, MSHS, HTI of tenth IHT period1、HTI2Delta HTI and correlation results of physiological indices measured in hypoxic chamber (n ═ 18)
A p-value < 0.05 (two-tailed) is considered statistically significant for the correlation analysis. SD represents sleep time (sleep duration); DSD denotes deep sleep duration (deep sleep duration); DSR denotes a deep sleep rate (deep sleep rate); DNA represents wake time (duration of night awaking); MSDS represents SpO while sleeping2Mean value (mean SpO)2during sleep); MHDS represents the mean HR reducing sleep mean value (mean HR reducing sleep); MSHS represents low oxygen segment SpO2Mean value (mean SpO)2 of hypoxic segments)
On the one hand, HTI was experimentally observed1Related to deep sleep time (r 0.515, p 0.029), Δ HTI and wake time (r 0.568, p 0.014). In addition to this, HTI1、HTI2Δ HTI and mean SpO2There was no significant correlation with the physiological indicators measured in hypoxic chambers. On the other hand, experiments have observed that three important indicators characterizing the risk of AMS are included: deep sleep time (r 0.492, p 0.038), deep sleep ratio (r 0.623, p 0.006) and average blood oxygen saturation (r 0.525, p 0.025). The effectiveness of DSI indicators in predicting AMS risk is therefore validated. Unfortunately, there was no significant correlation between the AMS predictor and the LLS indicator based on the data of the entire study population. One possible reason is that the LLS is judged based on subjective knowledge, judgment and willingness of the individual being evaluated. In addition, the least squares fit results of the correlation study are shown in fig. 5. These fitted curves show the data distribution trend well. A DSI contrast plot before and after intermittent hypoxic training (i.e., the first training period and the tenth training period) is shown in fig. 6. This figure illustrates that intermittent hypoxia training helps to improve the adaptive capacity of the individual's hypoxic environment.
When 12 training subjects with Δ HTI > 30 were studied in a centralized manner, the correlation between DSI and physiological measures in hypoxic chamber are shown in table 2, and table 2 shows the results of the analysis of the correlation between DSI measures of the intermittent hypoxic training sensitizers in this example and physiological measures in hypoxic chamber.
Table 2: correlation results between DSI of the tenth IHT cycle and physiological indices measured in hypoxic chamber (n 12)
A p-value < 0.05 (two-tailed) is considered statistically significant for the correlation analysis. SD represents sleep time (sleep duration); DSD denotes deep sleep duration (deep sleep duration); DSR denotes a deep sleep rate (deep sleep rate); DNA represents wake time (duration of night awaking); MSDS represents SpO while sleeping2Mean value (mean SpO)2 during sleep);MHDS represents the mean HR sleep time (mean HR reducing sleep)
The correlations between DSI and partial physiological indicators of the tenth training period are improved, including the deep sleep time (r 0.907, p < 0.001), the deep sleep ratio (r 0.783, p 0.003) and the average blood oxygen during sleep (r 0.730, p 0.007), DSI and LLS also have significant correlations (r 0.622, p 0.031), the least squares fit curve of the results is shown in fig. 7. Since HTI also has a role in predicting risk of AMS, Δ HTI reflects to some extent the individual's adaptability or sensitivity to hypoxia. Therefore, the proposed DSI index has better prediction effect on patients sensitive to intermittent hypoxia training (namely individuals with enhanced adaptability to hypoxia environment after training).
In summary, the above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A system for evaluating AMS risk based on IHT dynamic performance is characterized by comprising a data acquisition module, a data processing module, a robust enhancement module and an AMS severity monitor; wherein the content of the first and second substances,
a data acquisition module for acquiring SpO from multiple consecutive IHT periods2Information;
a data processing module for processing the acquired SpO2Processing abnormal values, filtering and dividing data segments of the information, and selecting low-oxygen data segments;
an index obtaining module, configured to obtain a dynamic blood oxygen saturation index DSI of the hypoxia data segment, where the DSI is SpO2Duration of holding above a set ratio, denoted as τi;
A robust enhancement module for the dynamic blood oxygen saturation indicator tauiCarrying out robustness enhancement to obtain an index taur;
AMS severity monitor based on a predetermined reference valueAnd said τrMake a comparison to satisfyThe people who train are qualified and suitable for entering the plateau area to meetPeople classified as non-adapted individuals are required to be further trained and observed, so that evaluation on AMS risk is realized;
the robustness enhancement is as follows:
ordering DSI sequence collected in one IHT period according to size, namely { tauiI 1.. lambda., lambda } where lambda represents the number of suitable hypoxic data segments when i is equal to 1<j is, τi≤τj(ii) a DSI acquired in one IHT period is defined as sequenceThe twentieth percentile of (d) is recorded as τr;
When { tauiWhen the base number of i ═ 1.. lambda. } is smaller than a set value, { τ } is estimated by the following methodiThe twenty-th percentile of i ═ 1.., λ }; the method specifically comprises the following steps:
let τ beiHas a maximum value ofMinimum value ofWill tauiRange of (1)Divided into sub-intervals of length l of delta segment
Wherein, pij(j ═ 1, 2.. multidot., δ) is τi1,2, λ falls in the jth subinterval, the twentieth locus falls in the subinterval z*Internal;
s.t.,g(z)≥0
whereinThe area of the first 20% of the total histogram area and the first z-1 segmentThe difference between the two;
let Pi00, the DSI of a hypoxic training period is defined as τr
τr:=τ+lτ
2. The system for evaluating AMS risk based on IHT dynamic performance of claim 1, wherein the outliers are processed as: and processing by adopting a method of replacing the abnormal value with a non-zero data average value before and after the abnormal value.
3. The system for assessing AMS risk based on IHT dynamic performance of claim 1, wherein the criteria for selecting the hypoxic data segment are: first, the length of the hypoxic data segment is greater than a preset valueminSecond, remove the average SpO of the last n samples2Hypoxic segments at levels greater than 90%.
4. The system for assessing AMS risk based on IHT dynamic performance of claim 3, wherein the preset/isminIs 250 seconds, and the value of n is 200 seconds.
5. The system for assessing AMS risk based on IHT dynamics of claim 1, wherein the dynamic oximetry indicator τ of the hypoxic data segmentiIs SpO2For a duration of more than 85%.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010264555.2A CN111513726B (en) | 2020-04-07 | 2020-04-07 | System for evaluating AMS risk based on IHT dynamic performance |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010264555.2A CN111513726B (en) | 2020-04-07 | 2020-04-07 | System for evaluating AMS risk based on IHT dynamic performance |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111513726A CN111513726A (en) | 2020-08-11 |
CN111513726B true CN111513726B (en) | 2021-04-13 |
Family
ID=71902298
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010264555.2A Active CN111513726B (en) | 2020-04-07 | 2020-04-07 | System for evaluating AMS risk based on IHT dynamic performance |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111513726B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113611414B (en) * | 2021-06-22 | 2023-03-28 | 北京理工大学 | Dynamic monitoring and performance evaluation method and system for acute altitude diseases |
CN113851224B (en) * | 2021-06-22 | 2022-06-03 | 北京理工大学 | Digital simulator for AMS risk assessment |
CN114464319B (en) * | 2022-01-13 | 2022-11-01 | 北京理工大学 | AMS susceptibility assessment system based on slow feature analysis and deep neural network |
CN115512849A (en) * | 2022-09-15 | 2022-12-23 | 北京理工大学 | Low-oxygen closed-loop intervention system of plateau pre-learning clothes |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019059366A1 (en) * | 2017-09-21 | 2019-03-28 | 京セラ株式会社 | Monitoring device, monitoring method, and monitoring program |
WO2019113562A1 (en) * | 2017-12-08 | 2019-06-13 | The United States Of America As Represented By The Secretary Of The Army | Systems and methods for monitoring subjects in potential hypoxic distress |
JP6670985B1 (en) * | 2018-10-26 | 2020-03-25 | 京セラ株式会社 | Monitoring device, monitoring method, and monitoring program |
-
2020
- 2020-04-07 CN CN202010264555.2A patent/CN111513726B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019059366A1 (en) * | 2017-09-21 | 2019-03-28 | 京セラ株式会社 | Monitoring device, monitoring method, and monitoring program |
WO2019113562A1 (en) * | 2017-12-08 | 2019-06-13 | The United States Of America As Represented By The Secretary Of The Army | Systems and methods for monitoring subjects in potential hypoxic distress |
JP6670985B1 (en) * | 2018-10-26 | 2020-03-25 | 京セラ株式会社 | Monitoring device, monitoring method, and monitoring program |
Non-Patent Citations (1)
Title |
---|
Intermittent altitude exposures reduce acute mountain sickness at 4300 m;Beidleman, BA; Muza, SR; Fulco, CS;;《CLINICAL SCIENCE》;20040331;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN111513726A (en) | 2020-08-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111513726B (en) | System for evaluating AMS risk based on IHT dynamic performance | |
Hovsepian et al. | cStress: towards a gold standard for continuous stress assessment in the mobile environment | |
Raoufy et al. | Classification of asthma based on nonlinear analysis of breathing pattern | |
Dougherty et al. | Comparing the detection of transdermal and breath alcohol concentrations during periods of alcohol consumption ranging from moderate drinking to binge drinking. | |
Wetten et al. | Relative validity of 3 accelerometer models for estimating energy expenditure during light activity | |
RU2623304C2 (en) | Monitoring system for patient monitoring and delirium detection | |
Nath et al. | Smart wristband-based stress detection framework for older adults with cortisol as stress biomarker | |
JP2013536971A5 (en) | ||
Skorucak et al. | Automatic detection of microsleep episodes with feature-based machine learning | |
Xu et al. | Identification of autism spectrum disorder based on short-term spontaneous hemodynamic fluctuations using deep learning in a multi-layer neural network | |
CN111508607B (en) | Obesity prediction system based on BP neural network | |
JP7262658B2 (en) | Systems and methods for camera-based quantification of blood biomarkers | |
Altini et al. | Cardiorespiratory fitness estimation using wearable sensors: Laboratory and free-living analysis of context-specific submaximal heart rates | |
US20230248320A1 (en) | Detection of User Temperature and Assessment of Physiological Symptoms with Respiratory Diseases | |
Tiwari et al. | Stress and anxiety measurement" in-the-wild" using quality-aware multi-scale hrv features | |
CN113903466A (en) | Data processing device and system for auxiliary evaluation of risk degree of cardiovascular diseases of population and application of data processing device and system | |
CN114464319B (en) | AMS susceptibility assessment system based on slow feature analysis and deep neural network | |
Li et al. | Automatic sleep staging by a hybrid model based on deep 1D-ResNet-SE and LSTM with single-channel raw EEG signals | |
CN113851224B (en) | Digital simulator for AMS risk assessment | |
Domingues et al. | Automatic annotation of actigraphy data for sleep disorders diagnosis purposes | |
Onim et al. | CASD-OA: Context-Aware Stress Detection for Older Adults with Machine Learning and Cortisol Biomarker | |
Intha et al. | Analysis of Effective Sleep through Logistic Regression | |
US20230284940A1 (en) | System and method for camera-based quantification of blood biomarkers | |
Rezki | Detecting Lie-A Practical Approach. | |
Austad et al. | Setting the Rest Interval in Actigraphy-A Systematic Review and Development of a Novel Manual |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |